| Time |
Nick |
Message |
| 01:05 |
|
olliy joined #minetest-dev |
| 01:05 |
|
DI3HARD139 joined #minetest-dev |
| 01:31 |
|
v-rob joined #minetest-dev |
| 01:39 |
|
fluxflux joined #minetest-dev |
| 04:02 |
|
behalebabo joined #minetest-dev |
| 06:31 |
|
calcul0n joined #minetest-dev |
| 08:00 |
|
ShadowNinja joined #minetest-dev |
| 08:01 |
|
Wuzzy joined #minetest-dev |
| 10:03 |
|
appguru joined #minetest-dev |
| 10:21 |
|
Kimapr joined #minetest-dev |
| 10:33 |
|
proller joined #minetest-dev |
| 10:51 |
|
Kimapr joined #minetest-dev |
| 11:31 |
|
appguru joined #minetest-dev |
| 11:54 |
|
Fixer joined #minetest-dev |
| 13:40 |
|
Kimapr joined #minetest-dev |
| 15:04 |
|
Kimapr joined #minetest-dev |
| 15:07 |
|
bebebeko joined #minetest-dev |
| 17:06 |
|
fluxflux joined #minetest-dev |
| 17:20 |
|
nashimus joined #minetest-dev |
| 17:28 |
|
appguru joined #minetest-dev |
| 17:45 |
Sokomine |
i have some trouble with latest git. visiting one particular place on a server makes the client extremly inresponsive - even when the fps are not yet down. fps also show a strange behaviour: they go continously up from i.e. 32 to 50 (quite rapidly but slow enough to see) - just to jump back to around 32 and go up again |
| 17:46 |
Sokomine |
another player reported that his fps where fine at that place but dropped considerably when i approached with my client. even tested with another account. the server is 5.2.0 |
| 17:48 |
Krock |
Sokomine: which server? |
| 17:48 |
Sokomine |
tunnelers abyss |
| 17:58 |
Krock |
Minetest hangs in the resend queue when the packets stall |
| 17:58 |
Krock |
2020-08-10 19:57:54: WARNING[ConnectionSend]: con(22/698) Packet quota used up for peer_id=1, was 1024 pkts |
| 18:00 |
Krock |
<Krock> likely already timed out server-side |
| 18:00 |
Krock |
likely already timed out server-side |
| 18:00 |
Krock |
the server flooded packets (maybe acks?) and the client spammed packets back afterwards. rip connection |
| 18:00 |
Krock |
<Krock> the server flooded packets (maybe acks?) and the client spammed packets back afterwards. rip connection |
| 18:01 |
Krock |
not sure what's happening there. looking for some network analysis stuff |
| 18:01 |
|
calcul0n_ joined #minetest-dev |
| 18:21 |
Sokomine |
yes, you timed out on the server. i wonder why there are so many packages? i do have a slow connection. but that should not affect other players |
| 19:07 |
Krock |
Excluded possible causes: mapblock mesh, active objects, node meta updates and delayed ACKs in the network |
| 19:07 |
Krock |
collision detection isn't the problem either |
| 19:08 |
|
search_social joined #minetest-dev |
| 19:16 |
|
wsor4035 joined #minetest-dev |
| 19:20 |
Krock |
neither minimap nor inventory updates. what's left open? network? I doubt that |
| 19:26 |
Sokomine |
extremly strange |
| 19:45 |
Krock |
problematic mapblock found (surprise!) might need to look into the inventory deserialization code again |
| 19:47 |
Krock |
or more general: profile MapBlock::deSerialize |
| 19:49 |
Krock |
Sokomine: for now, I'd suggest to remove machinery and special nodes with formspecs and check when the frequent updates stop |
| 19:51 |
Sokomine |
can you give me the coordinates of the block ingame? just saw the start of it. might even become tricky to remove any nodes there right now |
| 19:52 |
Krock |
add 16 in each direction |
| 19:53 |
Sokomine |
oh. so really only just one mapblock? strange. 16 in positive direction? |
| 19:54 |
Krock |
yes, the node causing the updates must be inside there |
| 22:08 |
|
Taoki joined #minetest-dev |
| 22:28 |
|
ghoti joined #minetest-dev |
| 22:59 |
|
proller joined #minetest-dev |
| 23:37 |
|
wsor4035 joined #minetest-dev |
